Denis Ndemwa

Inventing at the intersection of circuits, code, and intelligent systems.

A curated collection of my work in Data Science, Machine Learning, AI, and Business Intelligence.

🌐 Web Scraping

Extracted hockey team stats from ScrapeThisSite.

  • Python, BeautifulSoup, pandas
  • Automated multi-page scraping
  • Cleaned and exported data to CSV

🔗 Web Scraping Project ➜

🎬 Netflix Data Wrangling

Cleaned and prepared raw Netflix viewing data for analysis.

  • Python, pandas
  • Removed duplicates & missing values
  • Standardized dates and created new features (e.g., watch duration)

🔗 Netflix Data Wrangling ➜

📊 Titanic Exploratory Data Analysis (EDA)

Explored data to reveal trends, distributions, and relationships.

  • Python, pandas, matplotlib, seaborn
  • Visualized correlations and distributions
  • Detected outliers and data quality issues
  • Extracted actionable insights

🔗 EDA ➜

📈 Business Intelligence with Power BI

Built interactive dashboards for HR and sales data.

  • Power BI, DAX
  • Designed KPIs, slicers, and drill-down reports
  • Connected multiple data sources

🔗 Power BI ➜

📊 Data Visualization with Tableau

Created data stories and dashboards to highlight key insights.

  • Tableau, data preparation in Python/Excel
  • Designed clean and interactive dashboards
  • Published on Tableau Public

🔗 Tableau Project ➜

🧠 Interview with Geoffrey Everest Hinton (Godfather of AI)

A research project and summary of Geoffrey Hinton’s legacy in AI.

  • Collected highlights from interviews, talks, and papers
  • Focused on deep learning breakthroughs like backpropagation

🔗 Godfather of AI Project ➜

🔍 Regression Models

Built predictive models for continuous target variables.

  • Python, scikit-learn, pandas
  • Feature engineering & preprocessing
  • Hyperparameter tuning with GridSearchCV
  • Evaluated using RMSE and R²

🔗 Regression Models Project ➜

📂 Classification Models

Built machine learning models to classify data into categories.

  • Python, scikit-learn
  • Algorithms: Logistic Regression, Random Forest, etc.
  • Evaluated using accuracy, precision, and recall

🔗 Classification Models Project ➜

⚙️ MLOps

Explored deploying and managing ML models at scale.

  • CI/CD, Docker, MLflow
  • Built automated training pipelines
  • Experiment tracking and model versioning

🔗 MLOps ➜

✨ “Exploring data, building models, and visualizing the unseen stories behind the numbers.”